dora-rs

Mentors: Student: Ace20260322

This plan aims to improve dora-rs's getting-started documentation by creating a comprehensive learning system—including written tutorials, video guides, and runnable code examples—that enables new users to complete their first project within 2-3 hours. Leveraging prior contributions to the dora-rs ecosystem and experience producing AI-assisted programming tutorials, the applicant plans to deliver 8 milestones across 12 weeks: basic setup and rerun simulation before midterm (July 10), followed by multimodal model integration, LLM-based action planning, and adora+octos advanced features before the final deadline (August 24). All outputs will be published as open-source documentation and bilingual video tutorials on GitHub, Bilibili, and YouTube.

Mentors: Student: Gary Ding

Dora is a powerful robotics dataflow framework, but robot behaviors require fixed YAML graphs—no natural language interface, runtime error recovery, or high-level goal decomposition. This project builds an Octos agent bridge node for Dora, enabling natural language control of a simulated UR5e robot (e.g., "pick up the red ball and place it on the green plate"). The agent decomposes commands into DOT-graph pipelines, executing via dora transport tools, interfacing with dora-moveit nodes in MuJoCo. A prototype demonstrates pick-and-place with full step completion (vs partial completion with LLM-only reasoning). Deliverables: MuJoCo simulation, Octos bridge node, agent-callable tools, DOT-graph pipeline, mock provider, test suite, documentation. Stretch: Rust-native bridge, vision-based grasping, messaging channel integration

Mentors: Student: YujieSun

DORA lacks testing infrastructure. Debugging multi‑node dataflows means manual node isolation—slow and fragile. This project delivers `dora-test-utils`: a Rust crate that adds unit testing (NodeHarness), integration testing (TestSourceNode/TestSinkNode), and optional record/replay for regression. Deliverables include mock APIs, example pipelines, documentation, and CI templates. The work is scoped as a 175h MVP with a +175h extended version.

Mentors: Student: Dora—DGH

This project proposes dora-studio, a lightweight web-based GUI for viewing, editing, running, and monitoring Dora dataflows. Dora currently relies on CLI workflows, which can be less intuitive for onboarding, debugging, and demonstrations. The project will introduce a Rust-based control API to manage dataflows (list, start, stop, inspect) and a Vue-based frontend that provides a dashboard, a simple visual editor, and a live monitoring interface. The goal is to complete the full workflow loop: inspect, edit, run, and observe dataflows in a unified interface. Expected deliverables include the control API, a functional web UI for managing dataflows, a basic graph-based editor, live status monitoring, and documentation.

Mentors: Student: Shashwat Patil

This project aims to enhance robot data collection using affordable kits and the Dora framework. It involves building a modular MuJoCo simulation for robotic arms, implementing a Dora data collection node in the LeRobot format, and generating diverse manipulation datasets. The collected data will be documented, open-sourced on HuggingFace, and supported by beginner-friendly tutorials and demos to showcase real-world applications. Deliverables include simulations, a data collection tool, a sample dataset, documentation, tutorials, and a public dataset release.